Tuesday, February 19, 2013

M81 M82

M81 M82 by S Migol
M81 M82, a photo by S Migol on Flickr.
Via Flickr:
M81 M82 as seen on the night of Feb 16 just after midnight. I set up the camera and let it run from transit through until the fog came in at 3:30 am. This was in the backyard!

This is the result of a stack of 10 subs at 1200 seconds each for a total of 3 hours and 20 minutes of integration. There were some clouds that came through during the session so one frame was dropped in the middle of the sequence. The ISO setting for the camera was 100 ISO - something that I've been trying for a while to understand how to best use the low noise characteristics of this CCD based DSLR.


As usual, the imaging telescope is the Stellarvue SV4 with SSF 6 flattener and IDAS LPR filter.

What's changing is how I've been cooling the camera. I've taken apart the cooling system and started optimizing the way I'm using the TEC device. I've also started using a lighter-weight heat sink.

The camera temperature for this sequence was about 4C as reported by EXIF readings. The thermal probe on the cold finger was set to about -3 C and had reached equilibrium.

Guided with an SSAG on the SV 70 ED. I also disassembled the SSAG and hot glued the circuit board in place as I had read that some users saw movement when cables were pulled. I looked at mine and saw that it could move when prodded so I glued it into place.

Guided with Maxim DL with .5 second duration exposures. I know that this will chase seeing. I was rewarded with RMS values that were around .28 vs .33 pixel. FWHM values on the subs were correspondingly lower, too.

Calibrated with Maxim using 256 bias, 64 flats, and 17 darks. I also created a bad pixel map and have seen good results. I'll be tweaking these as I learn more about the concept.

Stacked and debayered with DSS.

PP in PixInsight:

DBE went well, used 2.5 .125 25 pixel size. Had to move several boxes due to bright stars and add another sample row along the edge. I should look at how the flats are being made.

Masked stretch to .3
Histogram stretch shadows .28 (about 1700 pixels clipped) midtones .14

Lightness extracted to make mask which was stretched and convolved for use.

a trous:
luminance
1 3,1,3
2 .01 1.5,.6,2
3 .015 1.5,.5,2
4 .01 1.2,.65,1
5 1,.6,1

for chrominance, ksigma on at threshold 2

Decon:
std 2.5
shape 2
aspect .95
rotation 35
deringing on .1 .1
5 layers

Classic star mask generated for MT Selection:
amount .5 selection .4

Back to luminance mask for Unsharp Mask
std 2.0 amount .65 on lightness L* deringing on

Asssited color calibration (no mask) to get rid of the remaining orange glow and muddy background from the sodium lights and pollution.
red .85 green 1 blue 1.15
saturation 1.5

Masked aggressive use of curves to drive background down and bring highlights up.

Here are the platesolve results:
Referentiation Matrix (Gnomonic projection = Matrix * Coords[x,y]):
-0.000016347086 -0.000530643527 +0.717252856801
+0.000530494575 -0.000016186406 -1.005633597950
+0.000000000000 +0.000000000000 +1.000000000000
Resolution ........ 1.911 arcsec/pix
Rotation .......... 88.236 deg
Focal ............. 655.19 mm
Pixel size ........ 6.07 um
Field of view ..... 2d 3' 15.4" x 1d 22' 17.9"
Image center ...... RA: 09 55 19.432 Dec: +69 23 51.50
Image bounds:
top-left ....... RA: 10 03 06.613 Dec: +68 22 50.96
top-right ...... RA: 10 03 08.084 Dec: +70 26 03.99
bottom-left .... RA: 09 48 14.218 Dec: +68 20 27.39
bottom-right ... RA: 09 46 46.531 Dec: +70 23 25.91



Screenshot while guiding